An agentic coding tool that works in your terminal.
Refact.ai
Open-source coding agent for VS Code and JetBrains.
SharkScore
#92 in Coding
Why it scores 35.3
Refact.ai has a SharkScore of 35.3/100, ranked #92 of 102 in Coding (updated 2026-11-18).
Each part compares Refact.ai with other listings in Coding. Evidence covers 87% of the score's weight.
- Adoption
- 31 · 30%
- User reviews
- no data · 25%
- Maintenance
- 16 · 20%
- Trust & transparency
- 60 · 15%
- Momentum
- no data · 10%
Paying never changes this score. How ranking works · method v2026-11-18.2
RankShark review of Refact.ai
Automated evaluation by RankShark · 18 Nov 2026 · method v1.0 · How we review
We were not paid for this review.
As of 2026-11-18, Refact.ai is a coding agent with self-hosting options, but its site says Refact Cloud has been discontinued and its repository is archived. [E1, MEASURED]
Refact.ai is a coding agent that the vendor says can work in IDEs, analyze a codebase, generate completions, and carry out coding tasks. The site also describes an on-premise option and says Refact Cloud has been discontinued. [E1]
Best for
Not for
- People who need Refact Cloud; the site says it has been discontinued. [E1] · home page
Key features
- Autonomous coding agent. The vendor says the agent plans and executes coding tasks, searches and analyzes repositories, and lets users preview and control the process. [E1] · home page
- In-IDE chat. The vendor says users can ask questions, edit, debug, and generate code in natural language. [E1] · home page
- Code completion. The vendor says completions use Qwen2.5-Coder and retrieval-augmented generation, with unlimited completions in the Free plan. [E1] · home page
- Tool connections. The vendor says the agent can connect with GitHub, PostgreSQL, Docker, databases, and CI/CD pipelines. [E1] · home page
- Self-hosting. The vendor says an on-premise version can run on the customer's own infrastructure. [E1] · home page
Pros
- The Free plan includes unlimited code completions and in-IDE chat with 32k context. [E1] · home page
- The vendor lists several IDEs, including VS Code and JetBrains, and offers an on-premise deployment option. [E1] · home page
- The vendor says users can bring their own key and connect models from providers including Gemini, Grok, OpenAI, and Deepseek. [E1] · home page
Cons
- The vendor says Refact Cloud has been discontinued. [E1] · home page
- Whether customer data is used for model training is not stated on the pages we checked.
- No release in the last 180 days (latest: release/v8.0.4 on 2026-05-14).
- The repository is archived by its owner.
Pricing
| Plan | Price | Limits |
|---|---|---|
| Free | Free | For personal and hobby projects; limited daily usage of the Autonomous AI Agent, in-IDE chat with 32k context, unlimited Qwen2.5-Coder completions, and a code-aware vector database (RAG). |
| Pro | From $10/month | For professional use; includes Free features, 40 Autonomous AI Agent requests per day, unlimited in-IDE chat with 64k context, and additional code-completion models. |
| Enterprise | Not stated | On-premise, SaaS, or AWS deployment; the page lists LLM fine-tuning, multiple-GPU optimization, code privacy, priority support, and onboarding. |
Checked 18 Nov 2026
- Free plan
- Limited daily agent usage; 32k-context in-IDE chat; unlimited Qwen2.5-Coder completions; code-aware vector database (RAG). · home page
- Free trial
- Not stated (checked 18 Nov 2026)
- Trains on customer data
- Not stated (checked 18 Nov 2026)
- Data retention
- Not stated (checked 18 Nov 2026)
- Data regions
- Not stated (checked 18 Nov 2026)
- Data processing agreement
- Not stated (checked 18 Nov 2026)
- AI models used
- Qwen2.5-Coder; Claude 4; GPT-4o; GPT-4o mini; and, with a bring-your-own-key setup, Gemini, Grok, OpenAI, Deepseek, and other LLMs. · home page
- API
- Not stated (checked 18 Nov 2026)
- SDKs
- Not stated (checked 18 Nov 2026)
- Webhooks
- Not stated (checked 18 Nov 2026)
- Platforms
- VS Code, JetBrains, Visual Studio (beta), Neovim (beta), Sublime Text (beta), PyCharm, WebStorm, GoLand, IntelliJ, and CLion. · home page
- Security claims
- The vendor says data is encrypted in transit using SSL/TLS. · privacy pageThe vendor says transmitted data is encrypted in transit and at rest. · privacy pageThe vendor says it uses role-based access controls and multi-factor authentication. · privacy page
What we measured
- GitHub stars · now
- 3,528
- Latest release · now
- release/v8.0.4 on 2026-05-14
- Tranco traffic rank · list of the day
- #968,575
- Repository · now
- archived by its owner
How we tested · sources
- Hourly uptime and response-time checks of the product's URL.
- Repository, release and download data from GitHub, npm, PyPI and Hugging Face where they exist.
- We did not sign up, pay, or use the product as a customer.
- home page: https://refact.ai/ · fetched 18 Nov 2026
- docs: https://docs.refact.ai/ · fetched 18 Nov 2026 · HTTP failed
- privacy page: https://refact.ai/privacy-policy/ · fetched 18 Nov 2026
- terms: https://refact.ai/terms-of-use/ · fetched 18 Nov 2026
- 18 Nov 2026: Plans or prices changed on the pricing page.
Something wrong? Report an error with a public link that shows it; corrections re-run the review.
Top alternatives to Refact.ai
See all- 74.5AgentPaid · from $17/mo
- 73.7
Open-source AI coding agent for the terminal.
AgentOpen source · from $10/mo - 71.7
Google's open-source AI agent for the terminal.
AgentOpen source - 71.3
Open-source platform for AI software development agents.
AgentOpen source · from $500/mo